Spectral Regression dimension reduction for multiple features facial image retrieval

Bailing Zhang*, Yongsheng Gao

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

3 Citations (Scopus)

Abstract

Face retrieval has received much attention in recent years. This paper comparatively studied five feature description methods for face representation, including Local Binary Pattern (LBP), Gabor feature, Gray Level Co-occurrence Matrices (GLCM), Pyramid Histogram of Oriented Gradient (PHOG) and Curvelet Transform (CT). The problem of large dimensionalities of the extracted features was addressed by employing a manifold learning method called Spectral Regression (SR). A fusion scheme was proposed by aggregating the distance metrics. Experiments illustrated that dimension reduced features are more efficient and the fusion scheme can offer much enhanced performance. A 98% rank 1 accuracy was obtained for the AR faces and 92% for the FERET faces.

Original languageEnglish
Pages (from-to)77-101
Number of pages25
JournalInternational Journal of Biometrics
Volume4
Issue number1
DOIs
Publication statusPublished - 2012

Keywords

  • Curvelet transform
  • Dimension reduction
  • Face image retrieval
  • Gabor feature/CT
  • LBP
  • Local binary pattern
  • Multiple feature fusion
  • PHOG
  • Pyramid histogram of oriented gradient

Fingerprint

Dive into the research topics of 'Spectral Regression dimension reduction for multiple features facial image retrieval'. Together they form a unique fingerprint.

Cite this